Exploring Reproducibility

Gabriel Mateus Bernardo Harrington

Research Associate | Webber Group

2024-01-19

Quick Questions!

Follow the link: https://www.menti.com/alqq45q4oprd

Discussion groups

Each table has a topic to discuss for 10 minutes
Please add your thoughts on the topic to the shared document: https://hackmd.io/@Coconut1019/BJJYscMY6/edit

Code Review

Join the team!
How does the community want it to be used?
- Quick second pair of eyes on code snippet?
- Methodological advice (QC cutoff, etc)?
- Full review of project codebase?

Code Review continued

Acknowledgements for review?
- GitHub has metrics
- Authorship? Establish group rules, or discuss ad-hoc?
- Minimal acknowledgement in GitHub Contributors file for reviewers?

Proposed workflow

Make a repo on DRI GitHub Org (perhaps using the template!)
Write and publish a registered report outlining the project
Gather/analyse data
1. Make ADDI workspace?
2. Use existing pipelines if applicable (Nextflow)
3. Containerise software environment (probably at end of project)

Proposed workflow continued

Publish all the things!
- Pre-print -> diamond open access
- Data on relevant public repositories + ADDI?
- generate DOI for code repo via Zenodo

Thanks for listening!

Here’s to making science slightly less crap!
Please get in touch about anything reproducibility related (or just just say hi :) - bernardo-harringtong@cardiff.ac.uk
Questions?

References